Journal of Medical Imaging最新文献

{"title":"Author-Centric AI Pre-Review: Interpreting Science Before It Is Judged.","authors":"Bennett A Landman","doi":"10.1117/1.JMI.13.2.020101","DOIUrl":"10.1117/1.JMI.13.2.020101","url":null,"abstract":"<p><p>The editorial explores an author-centric approach to AI in scientific publishing, arguing for the use of AI as a pre-submission self-review tool to help authors anticipate interpretation, clarify arguments, and strengthen rigor, while preserving author responsibility as well as the human core of peer review and research integrity.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 2","pages":"020101"},"PeriodicalIF":2.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13126655/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147822279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rolling convolution filters for lightweight neural networks in medical image analysis.","authors":"Naveen Paluru, Mehak Arora, Phaneendra K Yalavarthy","doi":"10.1117/1.JMI.13.2.024501","DOIUrl":"https://doi.org/10.1117/1.JMI.13.2.024501","url":null,"abstract":"<p><strong>Purpose: </strong>To introduce a filter design element called rolling convolution filters for developing lightweight convolutional neural networks (CNNs) in medical image analysis, aiming to reduce model complexity and memory footprint without compromising performance.</p><p><strong>Approach: </strong>Rolling convolution filters were generated by performing a channel-wise rolling operation on a single base filter, creating unique filters while restricting the learnable parameters. The method was applied to various two- and three-dimensional medical image analysis tasks, including reconstruction, segmentation, and classification across MRI, CT, and OCT modalities. The performance was compared with that of standard CNNs and other lightweight architectures.</p><p><strong>Results: </strong>The proposed rolling convolution filters substantially reduced the number of parameters and model size compared with standard CNNs, with a negligible increase in performance error. For quantitative susceptibility mapping, the rolling filter approach achieved results comparable to those of state-of-the-art methods with 6× fewer parameters. In COVID-19 anomaly segmentation, rolling filters performed on par with existing lightweight models while having <math><mrow><mo>∼</mo> <mn>68</mn> <mo>×</mo></mrow> </math> fewer parameters. For OCT classification, rolling filters maintained accuracy while significantly reducing the model size (49×).</p><p><strong>Conclusions: </strong>Rolling convolution filters offer an effective approach for designing lightweight CNNs for medical image analysis tasks, providing substantial reductions in model complexity and memory requirements while maintaining a performance comparable to that of larger models. This method can be easily incorporated into existing architectures and shows promise for deploying efficient deep learning models in resource-constrained medical imaging settings.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 2","pages":"024501"},"PeriodicalIF":1.7,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12956260/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147357113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ECLARE: efficient cross-planar learning for anisotropic resolution enhancement.","authors":"Samuel W Remedios, Shuwen Wei, Shuo Han, Jinwei Zhang, Aaron Carass, Kurt G Schilling, Dzung L Pham, Jerry L Prince, Blake E Dewey","doi":"10.1117/1.JMI.13.2.024001","DOIUrl":"10.1117/1.JMI.13.2.024001","url":null,"abstract":"<p><strong>Purpose: </strong>In clinical imaging, magnetic resonance (MR) image volumes are often acquired as stacks of 2D slices with decreased scan times, improved signal-to-noise ratio, and image contrasts unique to 2D MR pulse sequences. Although this is sufficient for clinical evaluation, automated algorithms designed for 3D analysis perform poorly on multislice 2D MR volumes, especially those with thick slices and gaps between slices. Superresolution (SR) methods aim to address this problem, but previous methods do not address all of the following: slice profile shape estimation, slice gap, domain shift, and noninteger or arbitrary upsampling factors.</p><p><strong>Approach: </strong>We propose ECLARE (Efficient Cross-planar Learning for Anisotropic Resolution Enhancement), a self-SR method that addresses each of these factors. ECLARE uses a slice profile estimated from the multislice 2D MR volume, trains a network to learn the mapping from low-resolution to high-resolution in-plane patches from the same volume, performs SR with antialiasing, and respects the image FOV during resampling. We compared ECLARE with cubic B-spline interpolation, SMORE, and other contemporary SR methods. We used realistic and representative simulations on human head MR volumes so that quantitative performance against ground truth can be computed. Specifically, healthy <math> <mrow><msub><mi>T</mi> <mn>1</mn></msub> </mrow> </math> -w and people with MS <math> <mrow><msub><mi>T</mi> <mn>2</mn></msub> </mrow> </math> -w FLAIR datasets were used for evaluations. We used the peak signal-to-noise ratio and structural similarity index measure as signal recovery metrics. We additionally used two independent brain parcellation algorithms, SLANT and SynthSeg, to compute the consistency Dice similarity coefficient and the <math> <mrow><msup><mi>R</mi> <mn>2</mn></msup> </mrow> </math> coefficient of determination, respectively, as comparison metrics.</p><p><strong>Results: </strong>For images with up to 5 mm of slice thickness and up to 1.5 mm of gap, ECLARE achieves greater mean PSNR and SSIM compared with other methods. In representative regions of interest, such as the ventricles, caudate, cerebral white matter, and cerebellar white matter, ECLARE performs comparably or better than other approaches. These trends are similar for both investigated datasets.</p><p><strong>Conclusions: </strong>The use of slice profile estimation, FOV-aware resampling, and self-SR allowed ECLARE to robustly superresolve anisotropic images without the need for external training data. Future work will investigate the utility of ECLARE on other organs, species, modalities, and resolutions. Our code is open-source and available at https://www.github.com/sremedios/eclare.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 2","pages":"024001"},"PeriodicalIF":2.3,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12959970/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147366995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Context-specific image quality assessment for virtual histologic staining: checklist and guideline.","authors":"Fadeel S Khan, Mia K Markey, Umberto E Villa, Alan C Bovik, Matthew C Fox, Brett H Keeling, Jason S Reichenberg, James W Tunnell","doi":"10.1117/1.JMI.13.S1.S11209","DOIUrl":"https://doi.org/10.1117/1.JMI.13.S1.S11209","url":null,"abstract":"<p><strong>Purpose: </strong>Virtual staining applies computational methods to transform optical images of biological samples into histology-like representations suitable for interpretation and analysis. Existing methods for evaluating virtual staining often prioritize metrics that do not provide a complete assessment of image quality for a given biomedical or scientific context.</p><p><strong>Approach: </strong>We review existing approaches to conduct image quality assessment (IQA) for virtual staining and identify their limitations. We make the case for context-specific IQA and propose a checklist and guideline for the comprehensive evaluation of image quality of a virtual staining system.</p><p><strong>Results: </strong>We present a context-specific IQA checklist and guideline for virtually stained images that (1) defines a specific context of use (COU), (2) explains the underlying mechanisms as they relate to COU, and (3) provides COU-specific evidence for validation. We build upon existing methods and connect IQA to the underlying imaging methodology and biological truth.</p><p><strong>Conclusions: </strong>The resulting checklist and guideline link engineering evaluation to biomedical utility and aim to enable more reliable development and evaluation of virtual staining systems.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 Suppl 1","pages":"S11209"},"PeriodicalIF":2.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13472531/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148766465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Synthesizing longitudinal ulcerative colitis progression via disentangled latent diffusion.","authors":"Umut Dundar, Alptekin Temizel","doi":"10.1117/1.JMI.13.S1.S11208","DOIUrl":"https://doi.org/10.1117/1.JMI.13.S1.S11208","url":null,"abstract":"<p><strong>Purpose: </strong>Synthesizing medical images at controllable disease stages while preserving patient-specific anatomy is hindered by the entanglement of pathological textures and structural features. We address this challenge for ulcerative colitis endoscopy, where severity follows a continuous ordinal progression along the mayo endoscopic score (MES).</p><p><strong>Approach: </strong>We propose disentangled anatomy-disease diffusion (DADD), a latent diffusion framework conditioned on a pretrained image encoder for anatomy and an ordinal embedder for disease severity. A feature purifier suppresses disease-correlated channels to produce cleaner anatomical representations and reduce semantic conflict between source anatomy and target pathology. These representations are injected into the denoising network through a triple-pathway cross-attention module with resolution-dependent routing. We further introduce delta steering, a training-free signal derived from ordinal embeddings that enables explicit single-pass control of disease transitions at inference.</p><p><strong>Results: </strong>On the labeled images for ulcerative colitis (LIMUC) dataset, the hierarchical DADD variant achieved a regression-judge quadratic weighted kappa (QWK) of 81.53 and root-mean-square error (RMSE) of 0.46, approaching the performance obtained on real images (83.28 QWK, 0.47 RMSE). It also achieved a Fréchet inception distance (FID) of 27.20 and improved downstream classification from 74.84% accuracy/83.14 QWK with real-only training to 77.75% accuracy/85.77 QWK when synthetic images were added.</p><p><strong>Conclusions: </strong>DADD improves controllable synthesis by disentangling patient-specific anatomy from pathological texture, enabling anatomically consistent visualization of severity progression and healing.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 Suppl 1","pages":"S11208"},"PeriodicalIF":2.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13481574/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148799781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Synthesizing raw multicoil MRI data with a deep generative framework.","authors":"Nikhil Deveshwar, Abhejit Rajagopal, Michael Lustig, Peder E Z Larson","doi":"10.1117/1.JMI.13.S1.S11211","DOIUrl":"10.1117/1.JMI.13.S1.S11211","url":null,"abstract":"<p><strong>Purpose: </strong>Deep learning-based magnetic resonance imaging (MRI) reconstruction can enable faster imaging and improved image quality; however, clinical adoption is limited by the lack of diverse raw multicoil training datasets. In routine clinical workflows, multicoil <math><mrow><mi>k</mi></mrow> </math> -space data needed for training are often discarded, and only magnitude images are retained. In addition, phase is not deterministically recoverable from magnitude, and storing full <math><mrow><mi>k</mi></mrow> </math> -space is impractical.</p><p><strong>Approach: </strong>We address these limitations by (1) using a deep generative model to produce multiple plausible synthetic phase images conditioned on a single magnitude image and (2) introducing a radiofrequency (RF) coil library to synthesize multicoil k-space from the generated complex-valued images.</p><p><strong>Results: </strong>The synthetic phase images resemble ground-truth phase as assessed by Fréchet inception distance, and the generated multicoil <math><mrow><mi>k</mi></mrow> </math> -space effectively trains unrolled MRI reconstruction networks.</p><p><strong>Conclusions: </strong>These results indicate that generative phase synthesis coupled with an RF coil library provides a practical alternative to real multicoil <math><mrow><mi>k</mi></mrow> </math> -space for training reconstruction models.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 Suppl 1","pages":"S11211"},"PeriodicalIF":2.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13675001/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How much of a face is a face: exploring reidentification potential with generative AI.","authors":"Chloe Cho, Yihao Liu, Bohan Jiang, Andrew J McNeil, Benoit M Dawant, Bennett A Landman, Eric R Tkaczyk","doi":"10.1117/1.JMI.13.S1.S11202","DOIUrl":"10.1117/1.JMI.13.S1.S11202","url":null,"abstract":"<p><strong>Purpose: </strong>Clinical photographs play an integral role across medical fields. Since the mid-20th century, deidentification has consisted of black bars covering specific facial features, typically the eyes alone. Although increasingly questioned, this practice persists in clinical and academic settings.</p><p><strong>Approach: </strong>A barrier to standardized deidentification guideline development is the unknown risk of artificial intelligence (AI) to reconstruct faces from partially obscured photos. We evaluate the ability of generative AI to reconstruct 10,000 facial images in the Synthetic Faces High Quality dataset across 14 regional masking strategies.</p><p><strong>Results: </strong>Covering the eyes or any other single facial feature resulted in highly identifiable reconstructions, demonstrated by low face mesh distortion (0.14 to 0.18 relative to whole-face masking; absolute total face mesh distortion 8.34 to 10.19) and high structural similarity index to the original face (1.24 to 1.25 relative to whole-face masking; absolute SSIM 0.91 to 0.92). An open-source face verification model using Dlib was able to match 97.98% to 99.93% of these reconstructed images with the original image prior to single feature masking. Removing all major facial features (eyebrows, eyes, nose, and mouth) resulted in a threefold reduction in face verification rates compared with eyes alone, from 98.87% (95% CI [98.63%, 99.07%]) to 33.93% (95% CI [32.95%, 34.94%]).</p><p><strong>Conclusions: </strong>We provide quantitative metrics of the reidentification risk that modern generative AI technology poses for partially obscured facial images.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 Suppl 1","pages":"S11202"},"PeriodicalIF":2.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12906867/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146208138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Image class translation: visual inspection of class-specific hypotheticals and classification based on translation distance.","authors":"Mikyla K Bowen, Jesse W Wilson","doi":"10.1117/1.JMI.13.S1.S11207","DOIUrl":"10.1117/1.JMI.13.S1.S11207","url":null,"abstract":"<p><strong>Purpose: </strong>A major barrier to the implementation of artificial intelligence for medical applications is automated CNNs' lack of explainability and high confidence for incorrect decisions, specifically with out-of-domain samples. We propose a generalization of image translation networks for image classification and demonstrate translation networks' potential as a more interpretable alternative to conventional black-box classifiers.</p><p><strong>Approach: </strong>We train an image-to-image network to translate an input image to class-specific hypotheticals, and then compare these with the input, both visually and quantitatively. Translation distances, the degree of alteration needed to conform to one class or another, are examined for clusters and trends, and used as a simple low-dimensional feature vector for classification.</p><p><strong>Results: </strong>On melanoma/benign dermoscopy images, a translation distance classifier achieved 80% accuracy using only a two-dimensional feature space (versus 85% for a conventional CNN using a <math><mrow><mo>∼</mo> <mn>62</mn> <mo>,</mo> <mn>000</mn></mrow> </math> -dimensional feature space). Visual inspection of rendered images revealed dataset biases, such as more scale bars in melanoma photographs than in benign lesions. Image distributions in translation distance space revealed a natural separation along the lines of dermatologist decision to biopsy, rather than between malignant and benign. On bone marrow cytology images, translation distance classifiers outperformed a conventional CNN in both 3-class (92% accuracy versus 89% for CNN) and 6-class (90% versus 86% for CNN) scenarios.</p><p><strong>Conclusions: </strong>This proof-of-concept shows the potential for image-to-image translation to go beyond artistic/stylistic changes and to expose dataset biases, perform dimension reduction and dataset visualization, and in some cases, potentially outperform conventional end-to-end CNN classifiers.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 Suppl 1","pages":"S11207"},"PeriodicalIF":2.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13349329/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148425162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"RadGazeGen: radiomics and gaze-guided chest X-ray generation using diffusion models.","authors":"Moinak Bhattacharya, Gagandeep Singh, Shubham Jain, Prateek Prasanna","doi":"10.1117/1.JMI.13.S1.S11205","DOIUrl":"10.1117/1.JMI.13.S1.S11205","url":null,"abstract":"<p><strong>Purpose: </strong>We present RadGazeGen, a framework for integrating experts' eye gaze patterns and radiomic feature maps as controls within text-to-image diffusion models to enable high-fidelity medical image generation. Although recent text-to-image diffusion models have achieved impressive success, textual descriptions alone often fail to capture disease-specific details necessary for generating clinically accurate and anatomically faithful images.</p><p><strong>Approach: </strong>To address these limitations, RadGazeGen leverages radiologists' eye gaze trajectories and radiomics feature descriptors as spatial and semantic controls in the diffusion process. Eye gaze patterns encode visuo-cognitive attention and spatial localization of subtle disease cues, whereas radiomics features capture subvisual phenotype characteristics such as texture, intensity, and shape. By combining these multimodal cues, the proposed framework guides the generative model toward anatomically consistent and disease-aware image synthesis. The quality of the generated images were evaluated using a board-certified radiologist.</p><p><strong>Results: </strong>RadGazeGen was evaluated on the REFLACX dataset for image generation quality and diversity. Furthermore, to assess its downstream clinical utility, the generated images were used for disease classification tasks on the CheXpert test set ( <math><mrow><mi>n</mi> <mo>=</mo> <mn>500</mn></mrow> </math> ) and for long-tailed learning evaluation on the MIMIC-CXR-LT test set ( <math><mrow><mi>n</mi> <mo>=</mo> <mn>23,550</mn></mrow> </math> ), demonstrating high fidelity and diagnostic relevance of the synthesized images.</p><p><strong>Conclusions: </strong>By jointly conditioning on gaze and radiomic representations, RadGazeGen bridges the gap between human visual cognition and machine perception, improving both realism and clinical validity in medical image generation. This framework underscores the importance of incorporating anatomically grounded and disease-aware controls in diffusion-based medical image synthesis.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 Suppl 1","pages":"S11205"},"PeriodicalIF":2.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13285692/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148310069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Importance of conditioning in latent diffusion models for image generation and super-resolution.","authors":"Dayvison Gomes de Oliveira, Franklin Anthony Ramos Coêlho, Thaís Gaudencio do Rêgo, Yuri de Almeida Malheiros Barbosa, Telmo de Menezes Silva Filho, Bruno Barufaldi","doi":"10.1117/1.JMI.13.S1.S11203","DOIUrl":"https://doi.org/10.1117/1.JMI.13.S1.S11203","url":null,"abstract":"<p><strong>Purpose: </strong>We investigate the use of latent diffusion models (LDMs) for synthesizing and enhancing photon-counting chest computed tomography (CT) images. We evaluate the models' capabilities in two main tasks: image generation for dataset augmentation and super-resolution (SR) for improving image quality, aiming to support diagnostic accuracy and accessibility to high-resolution data.</p><p><strong>Approach: </strong>The proposed framework combines a variational autoencoder-based latent encoder (AutoencoderKL) and a denoising diffusion model, trained under multiple conditioning tests. Eight experiments were conducted across generative and SR tasks, exploring the effects of different conditioning strategies, including segmentation masks and class labels (e.g., lung versus soft tissue), as well as varying loss functions.</p><p><strong>Results: </strong>Unconditioned LDMs produced hallucinated anatomy, lacking clinical interpretability. Conditioning with segmentation masks and anatomical labels considerably improved structural fidelity. The best results for image generation achieved a multiscale structural similarity index measure (MS-SSIM) = 0.7135 and peak signal-to-noise ratio (PSNR) = 24.53, whereas SR tasks reached MS-SSIM = 0.85 and PSNR = 27.31, comparable to recent diffusion-based benchmarks.</p><p><strong>Conclusions: </strong>LDMs show strong potential for both augmentation and SR of photon-counting chest CT images. When guided by segmentation masks and class labels, these models preserve anatomical structure and reduce hallucination risks. The results support their use in clinically relevant scenarios, providing controllable and high-fidelity image synthesis.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"13 Suppl 1","pages":"S11203"},"PeriodicalIF":1.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12904813/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146203298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}